© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 2768
    Prev
    Next
    INDEX
    Explanations

    increasing numerical values associated with specific attributes or parameters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    ngx_ types and c code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -------------</
    -0.69
     незавершена
    -0.68
    iſen
    -0.67
    iſten
    -0.66
     '\\;'
    -0.65
    alakip
    -0.63
    guiente
    -0.60
     Bewußt
    -0.60
    ьажоргаш
    -0.60
     ſind
    -0.60
    POSITIVE LOGITS
    t
    0.94
     t
    0.85
     T
    0.80
    T
    0.78
     s
    0.63
     int
    0.58
    getT
    0.57
    int
    0.56
     m
    0.54
     r
    0.51
    Activations Density 0.009%

    No Known Activations